If you ever look into a book on management, specially an American one, you'll probably see lots of impressive-looking diagrams, often labelled as "models". There are vague terms in boxes, and important-looking arrows connecting various important-looking boxes. Though the boxes are often labelled, the arrows hardly ever are. I find this irritating, because I'm usually more interested in the arrows than the boxes. I suspect most of these diagrams are practically meaningless - unlike the medieval emblems, which are didactic pictures where every little detail has a meaning.
I've always thought there could be a language of diagrams, a lingua franca of arrows, boxes, and shapes - much like computer flow charts, or CPM charts used in project management. When you had learned the language of diagrams (which need not be a highly complex one) you could glance at a diagram and quickly determine its meaning. The secret would lie in restricting the indiscriminate use of arrows, and having different types of line between boxes, to represent different meanings. The diagrams could be used to express research findings, theories, etc. There would be some words as well.
Though I haven't found any evidence of anybody ever designing such a language, a lot of work has been done on concept mapping in the last 10 years or so, and there are quite a few software applications that do different aspects of this. The commonest label for this seems to be concept mapping.
This page describes some of the concept mapping software that we've been trying out lately. It can be used at three points in an evaluation program:
1. When you are creating (or interpreting) a theoretical model to be tested through research;
2. When you have finished a qualitative study, and are trying to interpret the findings;
3. When you have interpreted the findings, and want to present them concisely.
Though any of the programs described below can be used at any of these three stages, in some cases it's just too difficult: why use Powerpoint to laboriously create a tentative mind-map (for point 2 above) when you could do it more quickly by hand?
Two web pages (that we know of) try to group this type of software:
For connecting concepts ("boxes and arrows" style), it's often easiest to draw a rough map on paper.This software performs a similar function, allowing you to link related concepts by drawing lines between them. The larger the diagram, the more time you save by using this software. We've been trying various concept mapping programs: Inspiration, Decision Explorer, Spidermap, CMap, and lately Omnigraffle.
Let's get the most primitive one out of the way first, mentioning it here only because so many people have this software. You may not know you have it, but it's built into Microsoft Office, and it comes up when you click on the Draw toolbar in Word, Excel, or Powerpoint. It's more powerful than it first seems, but it's more cumbersome to use than the other programs mentioned below. So I won't go into much detail on Microsoft Draw, except to offer a few general tips:
1. If you're making a boxes-and-arrows diagram, and you want to be able to move the boxes around, use Excel. The Draw menu looks superficially the same as with Word and Powerpoint, but if you look closely you'll notice that it adds another option: "Connectors." With these, when you move the boxes, the arrows move automatically - a big timesaver if you're trying to squeeze a lot of boxes onto one page.
2. A tiny movement of the mouse will often change the menu that comes up when you click on the right-hand mouse button, specially when formatting text boxes and finding graph options. Sometimes doing an exploratory right-button click is the only way you can know that an option exists.
3. If you need to make a lot of text boxes, and don't want the default format (12 point Times, big margins, black border), make one box first, customize its style, then make duplicates of that. Much quicker than styling each box individually.
4. Reading a book about MS Draw can save you a lot of time if you have a frequent need to produce complex diagrams. It has many functions that aren't at all obvious.
Inspiration is like an outlining program which does diagrams. Though that doesn't sound very impressive, it's one of the most user-friendly pieces of software I've ever encountered. Unlike many other programs, you can learn it quickly - so it's feasible to use on odd occasions. It's commercial software, but low-cost. You can label the arrows as well as the boxes,and flick between a diagram view and an outline view. My main complaint is that it doesn't let you superimpose additional graphics on a diagram. Printed output can look dotty, too - Inspiration needs to be able to output vector graphics.
Decision Explorer is another program that expresses concepts in diagram form - designed for working through decisions and their ramifications. Very powerful, but not suited to large chunks of verbatim text. One disadvantage of Decision Explorer is its very high cost, but a free demo version can be downloaded from the Banxia website. It's limited to 60 nodes, which is enough for many small projects.The demo version of Decision Explorer is included with the latest version of NVivo (which is described on our qualitative software page).
CMap comes from the University of West Florida. CMap makes it easy to create large boxes-and-arrows diagrams. Its great advantage is that, unlike other software of this type (except Inspiration - see above), it lets you name the lines as well as the boxes. This may not sound like much, but in fact it makes the software far more powerful for expressing thoughts clearly. This software, free for noncommercial use, has a wide range of uses apart from qualitative research. However, it can be quite annoying to use. Also, its archaic Windows 3 interface doesn't help, restricting file names to the DOS 8+3 character format.
Compendium (based on the earlier Questmap) is a still developing package, a little like CMap. At this beta stage, it's a free download.
Omnigraffle looks like the best yet. However it runs only on the Macintosh OS X operating system. Omnigraffle seems to do everything Inspiration does, combined with everything that MS Draw does - but we haven't yet tried it on a real job, and that's when you begin to find the problems with any software. A colleague who's tried it reports that it creates unnecessarily gigantic PDF files ot its maps. The Mac newsletter Tidbits, mid-February 2006, carried a rave review by Matt Neuburg - "I would be unable to convey in words how simple and clear it is to work in Omnigraffle. Everything about it is easy and delightful..." - with comments like that, from a reviewer who is often skeptical, I'll take another look.
Freemind is open source software for mind-mapping, available for Windows, OS X, and Linux. As it's only a mind-mapper, it's not as powerful as CMap or Compendium - but sometimes that's an advantage.
Though all these programs can be used to create influence diagrams they do slightly different things, in different ways. So it doesn't make sense to say that one is "better" than another: each is best for a specific purpose. For Audience Dialogue's most common purpose (helping to make sense of a mass of verbal data) none of them is quite right, but all are time-savers, with quite short learning curves.
"Arguments" here doesn't refer to pub brawls - more to the kind of polite disputes made famous by Socrates. It's about investigating claims that you might want to test in a research study - either to strengthen your own claims, or to challenge somebody else's (or vice versa, if you're really objective). And as for the word "analysing" in the above heading, the software itself can't do the analysis (yet), but makes it easier for you to do so by removing verbal clutter.
Claimaker and Scholonto
Claimaker is a spinoff from the Scholonto project from the Knowledge Media Institute at the Open University in Britain. (Scholonto is short for "scholarly ontologies" - but don't let that dismay you.) The Scholonto software, which seems to be the high-powered version of Claimaker, is software for turning an argument into a diagram, and closely examining the argument in that form. The diagram itself doesn't seem to contribute much, but the lack of words on it makes the argument clearer than if it is surrounded by a mass of verbiage. A free beta version is available for downloading.
How would you use this type of software in research or evaluation? Mainly at the beginning or end of a research report, when you are using (or preparing to use) the data you've gathered to support a particular line of argument. Perhaps there are recommendations that seem to flow from your findings - but how can those recommendations be justified? This is where software like Claimaker and Scholonto comes into its own: put your argument into a diagram, to reveal its weaknesses.